Visual Estimation of Independent Motions for 3D Structures in Dynamic Environments

Juan Carlos Ramirez and Darius BurschkaFaculty for Informatics, Technische Universitaet Muenchen, Boltzmannstr. 3, Garching bei Muenchen, Germany

ramirezd@in.tum.de, burschka@cs.tum.edu

Scene Tentative object candidates Encapsulated 3D blobs Motion estimation

An approach to consistently model and characterize potential object candidates presented in non-static scenes. Three principal procedures support our method:

i) the segmentation of the captured range images into 3D clusters or blobs, by which we obtain a first gross impression of the spatial structure of the scene,

ii) the maintenance and reliability of the map, which are obtained through the fusion of the captured and mapped data to which we assign a degree of existence (confidence value),

iii) the visual motion estimation of potential object candidates, through the combination of the texture and 3D-spatial information, allows not only to update the state of the actors and perceive their changes in a scene, but also to maintain and refine their individual 3D structures over time.

INTRODUCTION

Dinamic 3D Mapping

3D-MAPPING FRAMEWORK

3D-Blob Detection

After the supporting-plane detection, the 3D rigid registration is stored in an octree. In order to find the spatial relations among the 3D points a Depth-First-Search (DFS) is performed by transversing the leaves inside the octree and finally identifying and clustering the connected points.

VISUAL MOTION ESTIMATION

Plane detection and octree

Blob detection, clustering

Inliers, ego-motion

Outliers, independent-motion

Σ2=∑l=1

L

∥p2, l−(ego R⋅p1, l+ego t )∥

Cost function:

p1, j'

=hyp R⋅p1, j+

hyp t

The scoring is based on thesimilarity of matching points:

v jj=p2, j−p1, j'

χ j2=v jjS j

−1 v jjT< χ α

2

S j=P1, j+P2, j

EXPERIMENTS AND RESULTS

On a wheeled robot:

3D textured image

Static registrations andego-motion detection

Object-motion detectionin the map

Scene

In a table scene:

Object- and ego-motiondetection

Confidence value points(red) and object model

REFERENCES

VISAPP 2013 Barcelona, Spain, 21-24 February, 2013.

[1] Arun, K. S., Huang, T. S., and Blostein, S. D. (1987). Least squares fitting of two 3-d point sets. IEEE Trans. Pattern Anal. Mach. Intell.

[2] Kitt, B., Geiger, A., and Lategahn, H. (2010). Visual odometry based on stereo image sequences with ransac based outlier rejection scheme. In Intelligent Vehicles Symposium (IV), 2010 IEEE.

[3] Lin, K.-H. and Wang, C.-C. Stereo-based simultaneous localization, mapping and moving object tracking. In Intelligent Robots and Systems (IROS), 2010 IEEE/RSJ International Conference on.

[4] Moosmann, F. and Fraichard, T. Motion estimation from range images in dynamic outdoor scenes. In Robotics and Automation (ICRA), 2010 IEEE International Conference on.

[5] Nister, D., Naroditsky, O., and Bergen, J. Visual odometry. In Computer Vision and Pattern Recognition, 2004. CVPR 2004. Proceedings of the 2004 IEEE Computer Society Conference on.

[6] Ramirez, J. and Burschka, D. Framework for consistent maintenance of geometric data and abstract task-knowledge from range observations. In Robotics and Biomimetics (ROBIO), 2011 IEEE International Conference on.

[7] Wang, C.-C., Thorpe, C., and Thrun, S. Online simultaneous localization and mapping with detection and tracking of moving objects: theory and results from a ground vehicle in crowded urban areas. In Robotics and Automation, 2003. Proceedings. ICRA ’03. IEEE International Conference on.

[8] Wang, Y.-T., Feng, Y.-C., and Hung, D.-Y. Detection and tracking of moving objects in slam using vision sensors. In Instrumentation and Measurement Technology Conference (I2MTC), 2011 IEEE.